DOCSLIB.ORG

Good and Bad Properties of the Kelly Criterion∗

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Good and Bad Properties of the Kelly Criterion∗ Good and bad properties of the Kelly criterion∗ Leonard C. MacLean, Herbert Lamb Chair (Emeritus),School of Business, Dalhousie University, Halifax, NS Edward O. Thorp, E.O. Thorp and Associates, Newport Beach, CA Professor Emeritus, University of California, Irvine William T. Ziemba, Professor Emeritus, University of British Columbia, Vancouver, BC Visiting Professor, Mathematical Institute, Oxford University, UK ICMA Centre, University of Reading, UK University of Bergamo, Italy January 1, 2010 Abstract We summarize what we regard as the good and bad properties of the Kelly criterion and its variants. Additional properties are discussed as observations. The main advantage of the Kelly criterion, which maximizes the expected value of the logarithm of wealth period by period, is that it maximizes the limiting exponential growth rate of wealth. The main disadvantage of the Kelly criterion is that its suggested wagers may be very large. Hence, the Kelly criterion can be very risky in the short term. In the one asset two valued payoff case, the optimal Kelly wager is the edge (expected return) divided by the odds. Chopra and Ziemba (1993), reprinted in Section 2 of this volume, following earlier studies by Kallberg and Ziemba (1981, 1984) showed for any asset allocation problem that the mean is much more important than the variances and co-variances. Errors in means versus errors in variances were about 20:2:1 in importance as measured by the cash equivalent value of final wealth. Table 1 and Figure 1 show this and illustrate that the relative importance depends on the degree of risk aversion. The ∗Special thanks go to Tom Cover and John Mulvey for helpful comments on an earlier draft of this paper. 1 Good and bad properties MacLean, Thorp, Ziemba lower is the Arrow-Pratt risk aversion, RA = −u00(w)=u0(w), the higher are the relative errors from incorrect means. Chopra (1993) further shows that portfolio turnover is larger for errors in means than for variances and for co-variances but the degree of difference in the size of the errors is much less than the performance as shown in Figure 2. Table 1: Average Ratio of Certainty Equivalent Loss for Errors in Means, Variances and Covariances. Source: Chopra and Ziemba (1993) Errors in Means Errors in Means Errors in Variances Risk Tolerance* vs Covariances vs Variances vs Covariances 25 5.38 3.22 1.67 50 22.50 10.98 2.05 75 56.84 21.42 2.68 ### 20 10 2 Error Mean Error Var Error Covar 20 2 1 ∗ 100 u00(w) Risk tolerance=RT (w) = 1 where RA(w) = − Price (U.S. dollars) RA(w) u0(w) 2 15 900.00 14 PNP 13 800.00 % Cash S&P 500 Equivalent Loss 12 U.S. T-Bill 11 11 700.00 Means Nav Index 10 10 600.00 9 9 8 8 500.00 7 7 Index 400.00 6 6 5 5 300.00 4 4 3 200.00 3 S&P500 2 2 100.00 1 Variances Covariances 1 0 0 0.00 0 0.05 0.10 0.15 0.20 11/68 10/73 11/77 10/81 10/85 12/88 01 -02 -02 -02 02 -03 04 -04 04 -05 05 06 Magnitude of error (k) ar ar -03 -03 ar -04 -04 ar -05 -05 Dec- M Jun Sep Dec- M Jun Sep Dec- M Jun Sep Dec- M Jun Sep Dec- Dec- Date Figure 1: Mean Percentage Cash Equivalent Loss Due to Errors in Inputs. Since log has RA(w) = 1=w, which is close to zero, The Kelly bets may be exceedingly large and risky for favorable bets. In MacLean, Thorp, Zhao and Ziemba (2009) in this section of this volume, we present simulations of medium term Kelly, fractional Kelly and propor- tional betting strategies. The results show that with favorable investment opportunities, Kelly bettors attain large final wealth most of the time. But, because a long sequence of bad scenario outcomes is possible, any strategy can lose substantially even if there are many independent investment opportunities and the chance of losing at each investment 2 The Stochastic Programming Approach to Asset, Liability, and Wealth Management Figure 1.7. Mean Percentage Cash-Equivalent Loss because of Errors in Inputs Cash-Equivalent Loss (%) 11 10 9 8 7 6 5 4 3 2 1 0 0 0.05 0.10 0.15 0.20 Magnitude of Error (k) Means Variances Covariances Source: Based on data from Chopra and Ziemba (1993). Good and bad properties MacLean, Thorp, Ziemba Figure 1.8. Average Turnover for Different Percentage Changes in Means, Variances, and Covariances Average Turnover (% per month) 50 40 Means 30 Variances 20 10 Covariances 10 5 10 15 20 25 30 35 40 45 50 Change (%) Source: Based on data from Chopra (1993). Figure 2: Average turnover for different percentage changes in means, variances and co- variances. Source: Based on data from Chopra (1993) 12 ©2003, The Research Foundation of AIMR™ decision point is small. The Kelly and fractional Kelly rules, like all other rules, are never a sure way of winning for a finite sequence. In Section 6 of this volume, we describe the use of the Kelly criterion in many applications and by many great investors. Two of them, Keynes and Buffett, were long term investors whose wealth paths were quite rocky but with good long term outcomes. Our analyses suggest that Buffett seems to act similar to a fully Kelly bettor (subject to the constraint of no borrowing) and Keynes like a 80% Kelly bettor with a negative power utility function −w−0:25, see Ziemba (2003). See the wealth graphs reprinted in section 6 from Ziemba (2005). Graphs such as Figure 3 show that growth is traded off for security with the use of fractional Kelly strategies and negative power utility functions. Log maximizes the long run growth rate. Utility functions such as positive power that bet more than Kelly have more risk and lower growth. One of the properties shown below that is illustrated in the graph is that for processes which are well approximated by continuous time, the growth rate becomes zero plus the risk free rate when one bets exactly twice the Kelly wager. Hence it never pays to bet more than the Kelly strategy because then risk increases (lower security) and growth decreases, so Kelly dominates all these strategies in geometric risk- return or mean-variance space. See Ziemba (2009) in this volume. As you exceed the Kelly bets more and more, risk increases and long term growth falls, eventually becoming more and more negative. Long Term Capital is one of many real 3 6 Good and bad properties MacLean, Thorp, Ziemba Figure 2: Probability of doubling and quadrupling before halving and relative growth rates versus fraction of wealth wagered for Blackjack (2% advantage, p=0.51 and q=0.49). Source: McLean and Ziemba (1999) Figure 3: Probability of doubling and quadrupling before halving and relative growth rates versus fraction of wealth wagered for Blackjack (2% advantage, p=0.51 and q=0.49). Source: MacLean and ZiembaTable (1999) 3: Growth Rates Versus Probability of Doubling Before Halving for Blackjack Kelly Fraction P[Doubling before Relative of Current Halving] Growth Rate world instances in which overbetting led to disaster.Wealth See Ziemba and Ziemba (2007) for additional examples. ì 0.1 0.999 0.19 | 0.2 0.998 0.36 Thus long term growth maximizingRange investors| should0.3 bet_ Kelly or less.0.98 We call betting_ 0.51 for | 0.4 Safer 0.94 Less Growth 0.64 less than Kelly \fractional Kelly,"Blackjack whichí is simply0.5 a blend of Kelly0.89 and cash. Consider 0.75 the negative power utility functionTeams δ!|δ for δ <0.60. ThisRiskier utility function0.83 is concaveMore Growth and 0.84 when δ ! 0 it converges to log utility.| As δ becomes0.7 more_ negative,0.78 the investor is_ less 0.91 î 0.8 0.74 0.96 aggressive since his absolute Arrow-Pratt risk aversion0.9 index is also0.70 higher. For the case 0.99 δ of a stationary lognormal process and a given δ for1.0 utilityKelly function δw0.67and α = 1=(1 − δ) 1.00 between 0 and 1, they both will provide the same1.5 optimal portfolio when0.56 α is invested in 0.75 Overkill à 2.0 0.50 0.00 the Kelly portfolio and 1 −Tooα is Risky invested in cash. This handy formula relatingSource: the MacLean coefficient and Ziemba of (1999) the negative power utility function to the Kelly fraction is correct for lognormal investments and approximately correct for other distributed assets; see MacLean,In the next Ziemba two columns, and Li I will (2005). discuss For three example, topics: good half andKelly bad properties is δ = of− the1 Kelly log strategy and quarter Kelly is δ = −and3. why So this if led you me want to work a with less Len aggressive MacLean on path a through than study Kelly of fractional pick an Kelly strategies and appropriate δ. This formulainvesting does notusing apply unpopular more numbers generally. in lotto games For example, with low probabilities for coin tossing,of success where the returns where P r(X = 1) = p; P r(areX very− 1) large = q; (this p +illustratesq = 1, how bets can be very tiny) next month; and futures and commodity 1 1 1−δ 1−δ α α ∗ p − q p − q fδ = 1 1 = α α p 1−δ + q 1−δ p + q ∗ ∗ which is not αf , where f = p − q > 0 is the Kelly bet.
Load more

Recommended publications
  • Optimum Portfolio Diversification in a General Continuous-Time Model
    Stochastic Processes and their Applications 18 (1 Od4) Sl-?g 81 North-Holland OPTIMUM PORTFOLIO DIVERSIFICATION IN A GENERAL CONTINUOUS-TIME MODEL Knut Kristian AASE Telemark Distriktsh@gskole* Received 23 December 1982 Revised 8 September 1983 The problem of determining optimal portfolio rules is considered. Prices are allowed to be stochastic processes of a fairly general nature, expressible as stochastic integrals with respect to semimartingales. The set of stochastic different!11 equations assumed to describe the price behaviour still allows us to handle both the ass;r:ated control problems and those of statistical inference. The greater generality this approach offers cof~.xared to earlier treatments allows for a more realistic fit to real price data. with the obvious implications this has for the applicability of the theory. The additional problem of including consumption is also considered in some generality. The associated Bellman equation has been solved in certain particular situations for illustration. Problems with possible reserve funds, borrowing and shortselling might be handled in the present framework. The problem of statistical inference concerliing the parameters in the semimartingalc price processes will be treated elsewhere. portfolio selection * stochastic control * martingales * stochastic integrals 1. Introduction In this paper we consider optimum portfolio and consumption rules in a stochastic framework. We allow a fairly general model for the prices on assets, namely that the prices are semimartingales consisting of both continuous components and discrete jumps. In the time continuous case the geometric Brownian motion process has been a common model for prices in perfect markets. Also, more general, strong Markov continuous diffusion processes have been considered, like th2 Ornstein- Uhlenbeck process (see e.g.
    [Show full text]
  • The Kelly Criterion
    NORGES HANDELSHØYSKOLE Bergen, Fall 2012 The Kelly Criterion A closer look at how estimation errors affect portfolio performance by Adrian Halhjem Sælen Advisor: Professor Steinar Ekern Master Thesis in Financial Economics NORGES HANDELSHØYSKOLE This thesis was written as a part of the Master of Science in Economics and Business Administration at NHH. Neither the institution, the advisor, nor the sensors are – through the approval of this thesis – responsible for neither the theories and methods used, nor results and conclusions drawn in this work. 2 Summary The Kelly criterion are being used by many succesful investors and are often linked to their success in amassing great fortunes for both themselves and the funds they are running. This simple formula has made such a great impact on their results, that it is often cited as the main reason for these fund managers outperforming the indexes and other investors. However, there are several problems with using the Kelly criterion, and the main one involves the big risk that is being taken by sizing ones portfolios using the formula. This thesis will show that the problem with large risk becomes even greater when the errors in our estimation of the inputs start to get larger and larger. When our investor’s estimation errors increase, the volatility of the portfolio will not only increase as a result of this, but the growth of the wealth will fall. The conclusion that can be drawn from this is that risky assets, that is, assets that are very hard to determine the actual returns and volatility of with good accuracy, should use a fractional Kelly size ( ) when investing, to avoid overbetting on the asset, or even refrain from using the Kelly criterion at all.
    [Show full text]
  • Biased Roulette Wheel: a Quantitative Trading Strategy Approach
    Biased roulette wheel: A Quantitative Trading Strategy Approach Giancarlo Salirrosas Mart´ınez,CQF Department of Economics, Universidad San Ignacio de Loyola Abstract The purpose of this research paper it is to present a new approach in the framework of a biased roulette wheel. It is used the approach of a quantitative trading strategy, commonly used in quantitative finance, in order to assess the profitability of the strategy in the short term. The tools of backtesting and walk-forward optimization were used to achieve such task. The data has been generated from a real European roulette wheel from an on-line casino based in Riga, Latvia. It has been recorded 10,980 spins1 and sent to the computer through a voice-to-text software for further numerical analysis in R. It has been observed that the probabilities of occurrence of the numbers at the roulette wheel follows an Ornstein-Uhlenbeck process. Moreover, it is shown that a flat betting system against Kelly Criterion was more profitable in the short term. 1 Introduction [3], and statistical analysis [4],[5],[6]. While the core of martingale systems indicates the gambler to dou- Roulette wheel is one of the most famous games in the ble his bet in a consecutive manner after every loss context of casino gambling, although its invention has until a profit occurs which the following form: 20 · no clear author, it is meant to be introduced by the x; 21 · x; 22 · x; :::; 2n−1 · x thus, the cumulative loss is Pn i−1 french mathematician Blaise Pascal, in his efforts for i=1(2 )(x), where x is the amount of the initial bet building a perpetual motion machine around the 17th and n the number of bet the gambler has done.
    [Show full text]
  • Fortune's Formula
    Chapter 1 Fortune’s Formula In this chapter we will introduce the world’s most famous betting paradigm, Kelly criterion, often dubbed as the fortune’s formula. Initially used to exploit roulette game, later blackjack, then horse racing and finally stock market. 1.1 Coin toss Assume a simple coin toss game. We win if the coin comes up heads, if it is tails we lose. If we win, we receive og times the amount that we bet in addition to our original bet money. The definitions are: • p winning probability. • q loosing probability. • w number of wins. • l number of loses. • t = w + l number of all trials. • o odds, sometimes also called dividends. • og odds gain. Payoff from the winning bet in addition to the original amount. • b is the fraction of our wealth that we decide to bet. • W0 is our starting wealth • Wt is our final wealth after t trials 1 1. Fortune’s Formula Next please assume we play a slightly profitable game of coin toss. Prob- ability of winning the game is p = 0.4. Odds are o = 3.0, hence our odds gain is og = o − 1 = 2.0. We start with initial bank W0 of 100,- CZK and we bet b = 0.05 , 5 percent fraction of our bank every time. We play this game t = 3 times, we win some w = 2 and we lose some l = 1. Our final wealth W3 can hence be calculated as follows: 2 1 W3 = (1 + 2 · 0.05) · (1 − 0.05) · 100 (1.1) We make a gain of 2.0 times our decided “bank” fraction 0.05 twice and we loose 0.05 of our bank once.
    [Show full text]
  • Bayesian Kelly Criterion As an Allocation Strategy in Finnish Stock Markets
    Bayesian Kelly criterion as an allocation strategy in Finnish stock markets Jyväskylä University School of Business and Economics Master’s thesis 2020 Author: Risto Heikkinen Subject: Economics Supervisors: Juha Junttila & Juhani Raatikainen 1 ABSTRACT Author Risto Heikkinen Title Bayesian Kelly criterion as an allocation strategy in Finnish stock markets Subject Type of work Economics Master’s thesis Date Number of pages 10.11.2020 41 The Kelly criterion is an investment strategy, which aims to maximize long-term capital growth rate. There are known limitations in applying it’s original version such as high short-term volatility and uncertainty in estimating future returns. These limitations are traditionally tackled by risking only a fixed fraction of the proposed amount of the capital, but there is no clear consensus in published articles as how to choose this fraction. This thesis combines many earlier presented extensions to the Kelly criterion. These are e.g. estimating the uncertainty of future reaturns with a Bayesian model and a heavy tailed t-distribution and controlling short-term risk with a security constraint based on an investor’s risk tolerance. The end result is an algorithm which optimizes an appropriate allocation between a stock market index and a risk-free rate. In addition, there is a version of the algorithm, where it is possible to allocate the capital to one stock where the investor sees a special potential. In this thesis the potential is measured as the analysts’ target prices and the historical value of this target price information. The developed Bayes-Kelly method is applied to the Finnish stock market and it’s performance is studied during the years 2010 – 2019.
    [Show full text]
  • A Strategy of Maximizing Profit in the Long Run Using the Concept of Kelly’S Criterion
    Rochester Institute of Technology RIT Scholar Works Theses 5-3-2017 A Strategy of Maximizing Profit in the Long Run Using the Concept of Kelly’s Criterion Binyi Xu [email protected] Follow this and additional works at: https://scholarworks.rit.edu/theses Recommended Citation Xu, Binyi, "A Strategy of Maximizing Profit in the Long Run Using the Concept of ellyK ’s Criterion" (2017). Thesis. Rochester Institute of Technology. Accessed from This Thesis is brought to you for free and open access by RIT Scholar Works. It has been accepted for inclusion in Theses by an authorized administrator of RIT Scholar Works. For more information, please contact [email protected]. Rochester Institute of Technology A Strategy of Maximizing Profit in the Long Run Using the Concept of Kelly’s Criterion Binyi Xu A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science in Applied & Computational Mathematics School of Mathematical Sciences College of Science Rochester Institute of Technology Rochester, NY May 3, 2017 Committee Approval: Dr. Bernard Brooks Date Thesis Advisor Dr. James Marengo Date Professor Dr. Zhijian Huang Date Assistant Professor i ABSTRACT One of the fundamental problems of portfolio theory is how to rationally optimize the portfolio using diversification. In practice, maximizing the short term interest is not equivalent to maximizing the long term interest. Kelly’s criterion is considered to be the best strategy of maximizing profit in the long run. In this paper, we discussed the applications of Kelly’s criterion in various scenarios, including binomial cases, univariate stock, uncorrelated and corre- lated stocks.
    [Show full text]
  • Kelly Criterion
    Portfolio Choice and the Bayesian Kelly Criterion Sid Browne 1 Ward Whitt 2 Columbia University AT&T Bell Laboratories Original: March 4, 1994 Final Version: August 3, 1995 Appeared in Advances in Applied Probability, 28, 4: 1145-1176, December 1996 1Postal address: 402 Uris Hall, Graduate School of Business, Columbia University, New York, NY 10027 2Postal address: AT&T Bell Laboratories, Room 2C-178, Murray Hill NJ 07974-0636 Abstract We derive optimal gambling and investment policies for cases in which the underlying stochastic process has parameter values that are unobserved random variables. For the objective of maximizing logarithmic utility when the underlying stochastic process is a simple random walk in a random environment, we show that a state-dependent control is optimal, which is a generalization of the celebrated Kelly strategy: The optimal strategy is to bet a fraction of current wealth equal to a linear function of the posterior mean increment. To approximate more general stochastic processes, we consider a continuous-time analog involving Brownian motion. To analyze the continuous-time problem, we study the diffusion limit of random walks in a random environment. We prove that they converge weakly to a Kiefer process, or tied-down Brownian sheet. We then find conditions under which the discrete-time process converges to a diffusion, and analyze the resulting process. We analyze in detail the case of the natural conjugate prior, where the success probability has a beta distribution, and show that the resulting limiting diffusion can be viewed as a rescaled Brownian motion. These results allow explicit computation of the optimal control policies for the continuous- time gambling and investment problems without resorting to continuous-time stochastic-control procedures.
    [Show full text]
  • Risk Formulæ for Proportional Betting
    Risk Formulæ for Proportional Betting William Chin DePaul University, Chicago, Illinois Marc Ingenoso Conger Asset Management LLC, Chicago, Illinois September 9, 2006 Introduction A canon of the theory of betting is that the optimal procedure is to bet proportionally to one’s advantage, adjusted by variance (see [Ep, Th, www.bjmath.com] for discussion and more references). This is the well- known ”Kelly Criterion”. It results in maximum expected geometric rate of bankroll growth, but entails wild swings, which are not for the faint of heart. A more risk-averse strategy used by many is to scale things back and bet a fraction of the Kelly bet. This is done commonly by blackjack teams (see www.bjmath.com) and futures traders, e.g. [Vi], where the Kelly fraction is referred to as “optimal f”. In this article we examine what happens when we bet a fraction of Kelly in terms of the risk of losing specified proportions of one’s bank. We employ a diffusion model, which is a continuous approximation of discrete reality. This model is appropriate when the bets made are ”small” in relation to the bankroll. The resulting formulae are limiting versions of discrete analogs and are often much simpler and more elegant. This is the theoretical set-up used for the Kelly theory. The main result presented gives the probability that one will win a speci- fied multiple of one’s bankroll before losing to specified fraction as a function of the fraction of Kelly bet. This formula (2.1) was reported in [Go]. There it is derived from a more complicated blackjack-specific stochastic model.
    [Show full text]
  • Understanding the Kelly Capital Growth Investment Strategy Dr William T
    Investment Strategies Understanding the Kelly Capital Growth Investment Strategy Dr William T. Ziemba Introduction to the Kelly Capital Growth to finance, a number of note-worthy investors Alumni Professor Criterion and Samuelson’s Objections to it use Kelly strategies in various forms, including Sauder School of Business Jim Simons of the Renaissance Medallion hedge University of British Columbia The Kelly capital growth criterion, which fund. maximizes the expected log of final wealth, provides the strategy that maximizes long run The purpose of this paper is to explain the wealth growth asymptotically for repeated Kelly criterion approach to investing through investments over time. However, one drawback theory and actual investment practice. The is found in its very risky behavior due to the approach is normative and relies on the log’s essentially zero risk aversion; consequently optimality properties of Kelly investing. There it tends to suggest large concentrated are, of course, other approaches to stock and investments or bets that can lead to high dynamic investing. Besides mean-variance volatility in the short-term. Many investors, and its extensions there are several important hedge funds, and sports bettors use the criterion dynamic theories. Many of these are surveyed and its seminal application is to a long sequence in MacLean and Ziemba (2013). An interesting of favorable investment situations. continuous time theory based on descriptive rather than normative concepts with arbitrage Edward Thorp was the first person to employ and other applications is the stochastic portfolio the Kelly Criterion, or “Fortune’s Formula” theory of Fernholz and colleagues, see for as he called it, to the game of blackjack.
    [Show full text]
  • The Growth of Relative Wealth and the Kelly Criterion
    The Growth of Relative Wealth and the Kelly Criterion Andrew W. Lo∗, H. Allen Orr†, and Ruixun Zhang‡ This Draft: September 10, 2017 Abstract We propose an evolutionary framework for optimal portfolio growth theory in which investors subject to environmental pressures allocate their wealth between two assets. By considering both absolute wealth and relative wealth between investors, we show that different investor behaviors survive in different environments. When investors maximize their relative wealth, the Kelly criterion is optimal only under certain conditions, which are identified. The initial relative wealth plays a critical role in determining the deviation of optimal behavior from the Kelly criterion regardless of whether the investor is myopic across a single time period or maximizing wealth over an infinite horizon. We relate these results to population genetics, and discuss testable consequences of these findings using experimental evolution. Keywords: Kelly Criterion, Portfolio Optimization, Adaptive Markets Hypothesis, Evolu- tionary Game Theory JEL Classification: G11, G12, D03, D11 ∗Charles E. and Susan T. Harris Professor, MIT Sloan School of Management; director, MIT Laboratory for Financial Engineering; Principal Investigator, MIT Computer Science and Artificial Intelligence Labo- ratory. Please direct all correspondence to: MIT Sloan School, 100 Main Street E62-618, Cambridge, MA 02142–1347, (617) 253–0920 (voice), [email protected] (email). †University Professor and Shirley Cox Kearns Professor, Department of Biology, University of Rochester, Hutchison 342, Rochester, NY 14627, [email protected] (email). ‡Research Affiliate, MIT Laboratory for Financial Engineering, [email protected] (email). Contents 1 Introduction 1 2 Maximizing Absolute Wealth: the Kelly Criterion 4 3 Maximizing Relative Wealth 5 4 A Numerical Example 11 5 Testable Implications 14 6 Discussion 14 AProofs 16 1 Introduction The allocation of wealth among alternative assets is one of an individual’s most important financial decisions.
    [Show full text]
  • The Kelly Criterion Proceedings of the Business and Economics Section of the American Statistical Association, 1971 References 7-2
    The Kelly Growth Criterion Niels Wesselhöt Wolfgang K. Härdle International Research Training Group 1792 Ladislaus von Bortkiewicz Chair of Statistics HumboldtUniversität zu Berlin http://irtg1792.hu-berlin.de http://lvb.wiwi.hu-berlin.de Motivation 1-1 Portfolio choice Playing Blackjack Figure 1: 'Ed' Thorp Figure 2: Matlab GUI Motivation 1-2 Portfolio choice Wealth for discrete returns k Xi 2 R n 0 k 1 Y X Wn(f ) = W0 @1 + fj Xj;i A (1) i=1 j=1 + I W0 2 R starting wealth + I k 2 N assets with index j + I n 2 N periods with index i k How to chose fraction vector f 2 R ? Motivation 1-3 Managing Portfolio Risks Two main strands 1. Mean-Variance approach: Markowitz (1952), Tobin (1958), Sharpe (1964) and Lintner (1965) 2. Kelly growth-optimum approach: Kelly (1956), Breiman (1961) and Thorp (1971) Leo Breiman on BBI: Outline 1. Motivation X 2. Bernoulli - Kelly (1956) 3. Gaussian - Thorp (2006) 4. General i.i.d. - Breiman (1961) 5. Appendix Bernoulli - Kelly (1956) 2-1 Arithmetic mean maximization Consider n favorable Bernoulli games with probability 1 2 < p ≤ 1 (q = 1 − p) and outcome X = 1 (−1) For P(X = 1) = p = 1, investor bets everything, f = 1 n Wn = W02 (2) Uncertainty - maximizing the expectation of wealth implies f = 1 n X E(Wn) = W0 + (p − q) E (fWn−1) ; (3) i=1 Leads to ruin asymptotically n n o P (fWn ≤ 0g) = P lim (1 − p ) ! 1 (4) n!1 Bernoulli - Kelly (1956) 2-2 Minimizing risk of ruin Alternative: minimize the probability of ruin For f = 0 P (fWn ≤ 0g) = 0 (5) Minimum ruin strategy leads also to the minimization
    [Show full text]
  • Kelly's Criterion in Portfolio Optimization: a Decoupled Problem
    1 of 15 Kelly's Criterion in Portfolio Optimization: A Decoupled Problem Zachariah Peterson1,* 1Adams State University School of Business 208 Edgemont Blvd. Alamosa, CO 81101 [email protected] * Correspondence: [email protected] Abstract: Kelly's Criterion is well known among gamblers and investors as a method for maximizing the returns one would expect to observe over long periods of betting or investing. This paper will show how Kelly's Criterion can be incorporated into standard portfolio optimization models that include a risk function. The model developed here combines the risk and return functions into a single objective function using a risk parameter. This model is then solved for a portfolio of 10 stocks from a major stock exchange using a differential evolution algorithm. Monte Carlo calculations are used to directly simulate and compare the average returns from the Mean Variance and Kelly portfolios. The results show that Kelly's Criterion can be used to calculate optimal returns and can generate portfolios that are similar to results from the Mean Variance model. The results also show that evolutionary algorithms can be successfully applied to solve this unique portfolio optimization problem. Keywords: portfolio optimization; Kelly criterion; differential evolution; mean variance; logarithmic utility Key Messages: A decoupled form of logarithmic utility can be used to optimize portfolios The decoupled Kelly model generates portfolios with similar returns as the MV model The results were reached using test data with an intuitive solution JEL Classification: C6 Mathematical Methods; Programming Models; Mathematical and Simulation Modeling 1. Introduction In general, portfolio optimization problems aim to determine an optimal allocation of wealth among a pool of candidate securities.
    [Show full text]